The goal of this research is to determine the clinical value of high-resolution cell computer-based image analysis of urothelial cells in the sediment of voided urine. In prior studies, an objective classification of various classes of urothelial cells was achieved. An automated system of cell triage and classification has led to the construction of cytological profiles shown to be of diagnostic value in a test group of bladder cancer patients. These findings have been confirmed on a new cohort of 42 patients with a variety of conditions, such as bladder cancer and benign prostatic hypertrophy, with suitable controls. The sensitivity and specificity of this test run were excellent. This work has also led to an accumulation of morphometric data on urothelial cells that will provide additional diagnostic or prognostic information. We have been able to document differences between morphologically benign cells derived from patients with or without bladder cancer. This technique will also be used to determine whether there exist any subgroups of urothelial cancer cells that characterize patients at high risk for invasive carcinoma. The above studies were performed using scanning instruments that relied on a moving stage and photometer for image acquisition. We have since constructed a television scanning system that is capable of acquiring images at video speeds. Preliminary studies have shown that the cell images acquired with this system may be classified with an accuracy comparable with that from the prior systems. The system is also capable of performing DNA measurements on Feulgen-stained material, making it a useful tool in prognostic and diagnostic studies. (3)